Cardiovascular disease is the leading cause of death in industrialized nations. While interventional techniques such as angioplasty and/or stenting can often delay the need for coronary arterial bypass surgery (CABG), long-term efficacy still favors revascularization. Still today, 450,000 coronary bypasses are performed annually in the U.S. alone. Moreover, the number of revascularization procedures for peripheral arterial disease or hemodialysis access is increasing dramatically. Today, 400,000 hemodialysis patients in the U.S. need a patent vascular graft to survive. Creation and maintenance of hemodialysis access grafts takes up nearly 2% of Medicare's entire budget. In addition, population aging and increasing incidence of diabetes and obesity indicate that this is a major public health concern. A patient's own blood vessels clearly remain the best conduits for both coronary and peripheral bypass, as well as for hemodialysis access. Unfortunately, native vein or artery may not be available due to previous harvest or systemic disease progression. Indeed, the primary limitation to arterial revascularization is the availability of suitable native graft material. While synthetic blood vessels made from materials such as Dacron or expanded polytetraflouroethylene (ePTFE) perform well in large diameter applications, these synthetic conduits show unacceptably high failure rates in small diameter applications. We have recently reported excellent clinical trials results with a completely biological human vascular graft built in vitro using a self-assembly approach termed Sheet-Based Tissue Engineering (SBTE). However, a significant drawback of this self-assembly approach is the long production time. Here, we propose the development of a novel assembly strategy, called Thread-Based Tissue Engineering (TBTE). This novel approach offers the same advantages of a completely biological human product but will dramatically reduce the production time and cost. Combined with an allogeneic and/or devitalization approach, TBTE could finally make completely biological human vascular grafts available off-the-shelf and commercially much more competitive. In Phase I of this project, we demonstrated our ability to create threads of various strengths and sizes from both clinically relevant human cells and from canine cells. These threads were then woven into tubes on a custom circular loom to create human and canine vascular grafts. These grafts displayed promising mechanical properties in vitro. Phase I culminated in a short-term in vivo study that demonstrated the very promising clinical potential of human and canine grafts. These results met and exceeded all milestones set forth in the Phase I proposal. In Specific Aim 1 of Phase II, we will finalize the graft design based on promising data obtained in Phase I. These efforts will lead to six designs that will be tested in vivo in Specific Aim 2. Groups of four canines will receive H7 cm x 4.2 mm unendothelialized, devitalized grafts as arteriovenous shunts (femoral-to-femoral). The six designs will be: 1) Autologous, 2) Allogeneic, 3) Allogeneic decellularized, 4) Allogeneic gamma- sterilized, 5) Allogeneic seeded with autologous bone marrow in the operating room, 6) Allogeneic, punctured 3 times weekly with a 16Ga hemodialysis needle after a 3 month post-implantation maturation period. These groups cover the following cost-effectiveness range: 1<<5<<3<2<4. By comparing the performance of these groups, we can determine which design offers the best combination of efficacy and cost. Group 6 will give a first glimpse at the potential of this new type of graft to serve as a hemodialysis access graft. Finally, In Specific Aim 3, we will explore important parameters for the long-term commercialization of these grafts. We will study the use of serum-free medium to produce threads, which would improve reproducibility, reduce cost and simplify regulatory acceptance of the grafts (compared to bovine serum-containing medium). We will also study the use of molecular crowding additives to the culture medium. These large molecules have the potential to double or triple collagen assembly and hence, reduce production time and cost. Finally, we will determine if we can seed an endothelium on the woven grafts. While endothelium is not critical in high-flow applications like hemodialysis access grafts, it will be for other important markets/indications like coronary or lower limb bypass. However, establishing an autologous endothelium (allogeneic endothelium are highly immunogenic) introduces significant additional costs. As a cost-effective alternative, we will explore the possibility of replacing it by a new type of heparin coating. This Phase II project will provide: 1) the most cost-effective yet efficacious graft design; 2) initial in vivo efficacy data to justify a more extensive in vivo study to support and IND-submission; 3) additional data to further improve the grafts commercial and regulatory prospects.